Hi. In the previous video, we took a look at this data Rio Uni Site Programmer: It was originally about like, well, a typical configuration unit apparently was about thirty thousand dollars back in the day with all the extra you know cards and everything else. That's 1986 money. So this has one hideously expensive thing.

Anyway, look in the previous video. if you haven't seen the teardown of it, we're gonna have a go at repairing this. And now, if you remember the previous video, well I'll just show you we've got the leads over here, so just watch those leads there and we'll power it on. And I did a teardown the power supplies under here.

The power supply visually looked in good condition and everything else. Art was fine. so let me switch it on and I've got all the cards taken out just to minimize the load. so we switch it on.

the power LED comes on I think that one came up on and then just switched off and basically we measured it last time. There's test points here and there is no voltage on the test points, so it's like there's either something wrong with the power supply or what I think's more likely is the power supply is possibly shutting down because this is a ridiculously well engineered unit. So I'd Expect a similar sort of well and good engineering in the power supply in that there's no fuses loan in there and it obviously tries to power up. There's a lid coming on, it's doing something, but it's probably got detecting some sort of overload and shutting down, so that's the more likely scenario.

So there's basically two scenarios here. One is that the power supply. Big power supply under here. big beefy one is there's something wrong with that or B there's something wrong with this board here, or could even be this board over here because it's powered from the same rail.

ultimately. But if there's something wrong with the secretary on here, which is maybe shorting out a rail or doing something like that, So the first thing I'm going to do is well disconnect it and I'm gonna. so I think it's more likely that there's actually something shutting it down? I think there's more likely than not. So what I'm gonna do is measure the measure the rails.

Here we go. Let's give it a ball. We've got five volt rail. we've got a 12 volt rail, a 48 volt rail minus 5, and my and 2.1 So let's go ground and let's go 5 volts first because that's the biggie because all the chips on this board hundreds of them.

I Don't have this. hundreds. it's probably 50 or something five volts. So if one of those is shortened or one of the caps I Mean you know there's each one of these chips is going to have one of these little axial bypass caps.

These caps here actually look in good. Nick There's the vent holes. There's no bulging in the vent holes. look good.

I Don't think there's any Tanner Lumps on this board. By the way, a simple repair thing. Yes. I've done the visual like I've looked around, everything's fine.

You do the smell test, everything's fine, so there's no visual signs of any problems. So maybe we've got a a cap shorted out or something. So let's have a look at the rails. Hello, hello, point two and it's not charging up point one if anything is dropping.

So what you do here is swap the probes, see if there's some no our five volt, our five volt rail is shorted Wow That's it. Ah, yep, that was the most likely scenario, but it looks like it's paid off. So how five volt rail was shorted great, but got a we've got something to work from be it explains the symptoms of the power supply shutting down. So I think the power supply is probably fine and it's just attacked in that short and shutting the thing is shutting down all the rails just to protect the whole instrument.

so that's great. But as I said yeah, we've got LED bypass caps on every single chip. So what we need is a high resistance, high resistance, high resolution multimeter that can go down to one milli ohm or thereabout. So evening, start tracing down a short.

Actually, let me let me measure the other I'll start go off half and well, you know we found a different problem. Okay, so there you go. 12-volt rails fine, the other rails fine, Whatever that is. 48 volts minus 5 volts.

Yep. two point one volts. Yep, and yep. So all the others good.

5 volt rallies Cactus. All right, let's chase this one down the rabbit hole. a meter with one milli ohm resolution. and let's measure that again.

Okay, two two eight, one two two six two two five. whatever you want to call. it's like going down slightly and let's go somewhere else on the board. Let's go all the way over here.

there's your another cap. Oh ooh, that's no good. So much for narrowing this sucker down. That big ground plane in there is gonna ruin our day.

What if I keep one probe on there one over here? Okay, yeah, here's the problem. We're gonna have to get better resolution than this and or like, yeah, we can like short out the probes. You know you can do your relative thing. If you've only got this, where is that null like that? you know and you can go.

This is where you need good sharp probes. Okay, there you go. 186 187 to the short Oh 176 to the short. Where else I don't know? there's another cap all the way over here.

You need good sharp probe. So I'm using my probe master once with the ridiculously sharp tips on them so I can penetrate the oxide and really get in there. And well, this is not great. We're gonna maybe I need more a higher resolution bench meter to try and tackle this one.

It wouldn't be the first time that I've had to resort to a bench meter like a six and a half digit bench meter to get the resolution required to trace down a short on the board. Check this out. I Thought I'd take the board out just so it's you know, easier to work on and maybe isolate one of the other sections. And this five volt rail? 56 Ohms now not the same in the other direction.

Give me what I range in 79. Ohms There you go. So the short is not actually on this board. Now that's you know that wouldn't be uncommon I wouldn't quibble about that for a fight.

you know, a five all rail with a huge number of TTL Well, you know CMOS ETTL type chips for this sort of thing. So I deem that to be okay. So I reckon the short somewhere else. Hmm.

narrowing it down I'm actually glad it's not on here. so process of elimination will plug in the memory expansion board there, so look Tong at the right angle. Hey Nope, it ain't there. All right floppy drives.

Let's plug in the floppies. Is it the poor old floppy drive? Nope. it's got to be the backboard. that's the only one left.

Ha. So just probe the I don't know what pins are down there. I'll just probe pin 10 and 20 on a chip here. that'll be the 5 volt ray up.

Bingo! Found it definitely short on the backboard. so I need to unscrew that, get it out and then we can work on that separately. Don't want to work on it when it's in the back there? That's alright. Tell you what, that's not bad design.

There's only one screw top and bottom there and then this entire piece comes out with these um, little bracket little hooks in there to hook into the bottom. Wow, that's nice. What the hell is? Wow Well, the heatsink device is on there. It's going to say that had nothing to do with the power supply, but that's the waveform generator board.

so that's the external heat sink. It's all passive. of course. there's no fan in that it's all the passive heat sinking for the drive.

There's all the drive transistors down in there for the driving, the VCC of the pin, and and the programming pin. So this is the waveform generation board as we went through in the previous video. but wow, that's a so beast of it. He seemed think these things are bloody huge power amplifier or something to program a little piddly chip.

Wow, talk about over-engineered Now the first thing I see I'm not sure if they're across the rails or not, but all these old-school take 10 alums and yeah, they're one of the first culprits that you'd suspect. but I don't know if they're just across the five volt rail or not. But once again, you know you give this thing a visual check. Make sure nothing is dodgy there.

but no, it looks ok. but you know it could be a chip shorted. it could be a cap. You know it's unlikely to be like a PCB fault or something like that in a bit of gear that made it into production.

You know, like a short on internal layer or something like that. So yeah, it's got to be some sort of Kapil or our component I Tell you what, we may not need our high resolution meter. After all, we might be able to do this I've knowed this out I Just know. near enough.

Yep, good enough for Australia The first thing I did is I don't have? well I do have the schematic for this board so I could look it up, but I'm going to try and do it without it. So I went through and measured or across all the Tanum caps and the only one that was across the rail was this one here. and well, that was across the five. Ah, trails all the others.

There we go. it's 55 milliamps right? And then if we go up here and measure that same chip as measuring before, no, sorry, that's the analog devices part that's not the correct pins. There you go: 211 milliohms so you can see that's higher resistance over here and we can measure a much lower value here. So it's got to be closer to this area than it is to this because it's got.

If let's assume that that cap is shorted for example, let's assume that's the culprit it could very well be. Then you know if you measure any two points on opposite sides of the board, they should be higher resistance over here. And that's how you can narrow it down. You can sort of.

you know, measure resistance on points and and things if this one actually doesn't have too many 5 volt digital chips. Actually, because this is the waveform board, it's got just tons of other stuff. So, but that one is like that's as low as you get I don't see anything else digitally around there. So really I'm you know I would almost suspect that I would suck that one out.

Really? Maybe I'll find another digital chip over here somewhere. Yep, sure enough, check this out, right? And this one over here is 208 milliohms and this one over here which is a HC 373. it's kind of further I don't know? kind of. Well, you could say similar similar distance and bingo, that's 221 So that's slightly higher so you know if the shop was here of course then it just doesn't make sense.

The short has to be within this area down here, so I reckon that little tailor a little tag. tantalum. evil little suck as they are. they can develop our shorts and they can catch a light and explode with pulse input.

You know, excessive pulse currents and things like that? Anyway, I'm gonna suspect that sucker and disorder that one. I think because I can't really find anything else around there. Five volts. I don't want to go consult the manual and I don't I don't have the board overlay but I've got the schematic.

but yeah, that's good enough evidence. and as further confirmation is puppy up here there we go. Three hundred millions so that is further away than there. So yep, that makes sense and I've sucked it up straight across that cap again.

Bob's your uncle? Look at that. No workers. haha yep and haven't measured this yet. Go ahead that third hand, but I'm sure if I do that.

Come on little turd. Bingo tag. Tantalum classic. Absolutely classic Culprits: these things: Pain in the ass? look at that 10 Mike 50 volt job.

STC For those playing along at home who want to run the numbers, let's go between this, the ground yep, the ground there. There you go. So that's 67 million ohms and the positive. Over to our reference: I don't do I use that one of that one doesn't matter over to our reference chip 163 and you can probably see if we go over to this one over here.

see it's higher and that's just the resistance going across the board across the ground plane. Plus the you know, the thermal relief, small traces going into there and the joints and everything else. So here you go: Beautiful and it's time to get out there. Gorgeous! AVX Sample kit that I got.

this was one of many I got in the mailbag, thank you very much AVX Tantalum sample kit of course, don't have any of that newfangled tagged tantalum crap these days. Um, but there's a 10 Mike 50 volts. even though we don't need it like there's no reason that like. it's just impossible that you could possibly get 50 volts across that cap because it's directly across measured across the 5 volt rail.

I Don't even have to check the schematic so you know. Anyway, I'll go with the same one. So there you go, we'll use one of those. have to hold my tongue at the right angle.

What? There we go. Um, even though it's a surface mount job either. So the pin pitch that looks very similar. oh just Bojan an SMD one? Yeah, she'll be right.

no worries, that looks like a bought one. Let's go. Alright, let's power this puppy back up. I've measured the 5 volt ray.

oh it's 11 ohms in both directions so that sounds pretty good. It's give her a bull and here we go. Green lid for lids If I hear the floppy drive go I might might wet my pants. sweet.

Not often do I get one that basically wisdom wound. Still like we've fixed that fault. Whether or not there's other faults I don't know, but we're certainly getting progress. It's not very often that I get a repair like this that pretty much comes down to exactly what I thought it would be I thought it was.

you know most likely to be not the power supply but actually short on the board shutting down the power supply cuz that all made sense from an engineering perspective and it turns out yep it was a short on the board and then I'd been narrowed it down, got down to that board and then hey tag tan alums yeah let's check those and sure enough one of those babies confirmed. confirmed with multi point measurement on there pretty much you know triangulated kind of thing and that was it. Bob's your uncle that's a Bobby does tha well that so we've got some LEDs could happen I need to make boot floppies and everything like that. All right.

So I put all the cards back in and we pair it up and this led over here. It's not very bright cheese, you can't see it. but anyway, power and self-test there are on I Believe the self tester takes quite some time so I'm going to leave that running. consult the manual how to use this thing and go find a D25 serial cable to hook up to this thing and you may not I can't remember, but you may not need the floppies to actually you know, at least get something out of the serial port to begin with.

The ROM may actually do that. You know it may not. It may tell you, you know, no load the boot put in the boot disk or something like that. So if we can get that far, if we can get something out of the serial port.

Well, it's been over half an hour now and the self test lead is still lit. So um, which is like it's supposed to do this when you pair it on power and the self-test letter supposed to be lit and then it doesn't say how long the self-test takes but Secret squirrel told me it was like yeah, 20 minutes or something like that so it hasn't gone off. so this is not good. Hmm and you know they're supposed to be like blink.

Once it's finished it's self test. It'll like blink and tell you any errors or anything like that. So yeah, it's not great. By the way, I Found out what this was on the roms here Property of Copyright 1983 H and I Inc H Anna actually stands for Hunter and Ready.

My mate Steve leaps and actually I commented on the video and he knew all about this back in the day and apparently Hunter and Ready they developed basically. The first are theis the first real-time operating system I believe it was called via T X and that's probably what's running inside here, hence the copyright. Thank you very much! Steve We've interviewed Steve on the amp hour and that was very interesting I have to leave that one in down below. Check it out, right? So Ivory powered it with the X memory board disconnected and same thing the LED is still on.

So yeah. hmm. anyway. I Forgotten all the excitement of our repairing this thing, forgot to measure the voltage as I've measured them all and all the test points are fine.

so all the rails are up, so it's not that all right trying to get the scope out and see if this thing does anything. I'm using the terminal mode here. pin 2 is the transmit in DTE mode data terminal equipment pin 7 of the ground and if we switch it on here, we get our regular Rs-232 Y levels. There were like - you know, five and a half volts or something like that and not a sausage.

But check this out so well. obviously it doesn't output anything until it's finished the self-test and it's not finished in the self-test But if we switch it off, tada, look at that. So let's actually try and capture that, shall we? So here we go. Bingo! We've actually got some data there you go.

so it is spitting something out when you switch it off, it's like not on shutting down or something like that. Hmm. Anyway, at least is doing something shows that. the process is working and everything else.

And sure enough, if we actually capture and decode that 9600 board, we can actually see what it says that's carriage return and then line feed, Line feed, line feed and then you can see on the list of here Power Space Down Awesome! So it's powering down. thanks for that Uni site and I managed to find an old school ID 25 - Dean I and adapter. I've got a cable run over the PC and we can simply use a terminal emulation program to check it out. Don't need the scope and the dodgy wires anymore.

Sure enough, there it is that's Power Up. We just get nothing. There's a whole bunch of line feeds in there and we just get the message Power Down Well, at least it's doing the right thing and actually gracefully shutting that down like a soft power switch because I'm physically doing the hard mains power switch on the back so it's obviously detecting those power down. It's got enough power reserves and it's simply doing the right things: shutting down the OS, doing whatever it needs to do, and outputting the serial commands.

so it's obviously still got the power reserves to actually keep that process of going and doing that. So there's got to be some sort of power down watchdog interrupt type thing and I'm sure if you check the schematic it's in there. It interrupts the processor and goes what? go to your power down routine before we actually lose power. Awesome! So unfortunately that's going to be it for this episode anyway.

we have actually repaired the thing to the point where it's booting up and it's doing its you know, given us the proper data out, but the rest may like, you know, some sort of software or configuration problem or something like that which needs a different class of troubleshooting. I'm afraid. so I'm gonna have to leave that to a second video. but if you have any ideas if you've seen this problem before, if you're familiar with it where it just sticks continuously in the self-test mode, then let me know.

I've as I said, I've tried with them without the memory card and stuff like that and that. I put it like a blank floppy in the drives and it doesn't. you know, seem to make a difference it and it doesn't matter how long I leave this thing, it just sits in self-test mode there and never exit. It's the letter supposed to turn off when it's finished or given error code and start blinking.

LEDs and things like that and nothing comes out of the serial port after boot up. So yeah, it's a bit disappointing. Isn't the cheese? I Hope I would have yes and I've plugged in the pods to by the way and it doesn't make a difference there either. so very strange.

So anyway, I hope you found that repair interesting and useful. If you did, please give it a big thumbs up. As always, discuss down below: links subscribe videos at the end. Here somewhere are the ones to watch.

Catch you next time.